qdrant · Issues· 731 open

[FEATURE] Surface prefetch component scores (BM25, cosine similarity) as input to reranking pipelines

Flaky test `model_testing::tests::harness`

Flaky test `segment_builder_test::test_building_cancellation`

Flaky test `continuous_snapshot_test::test_continuous_snapshot`

Nightly model testing failure

Flaky test `model_testing::tests::harness_no_restarts`

Flaky test `tests::wal_recovery_test::test_wal_replay_loads_pending_to_queue`

Flaky test `tests::deferred_points_tests::test_deferred_points_wait_false`

Context Search returns negative loss scores inconsistent with the documented formula

Fusion inside a prefetch is computed per shard, so its scores and score_threshold depend on shard_number

Flaky test `model_testing::tests::harness_no_optimizer`

fix(uio): async disk-cache open panics on Populate::Partial for zero-length objects and empty ranges

Title: is_empty filter behaves differently depending on whether the field is indexed

Binary-quantized Euclidean query results are not ordered by reported score when rescore=false

Query returns zero rank at the documented equality boundary